INDEX

Explanations

mentions of discrimination based on various factors such as race, sexual orientation, and nationality

oai_token-act-pair · gpt-3.5-turbo

references to discrimination in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Interstitial

-0.86

sis

-0.72

ski

-0.71

Adds

-0.70

zyme

-0.68

ECD

-0.67

orah

-0.67

tom

-0.65

bold

-0.65

Sed

-0.64

POSITIVE LOGITS

 discrimination

1.22

 Discrimination

1.03

 prejudice

0.96

 discriminated

0.94

yip

0.91

rimination

0.87

 discriminate

0.87

discrimination

0.85

 prejudices

0.84

 protections

0.81

Activations Density 0.016%

mentions of discrimination based on various factors such as race, sexual orientation, and nationality

references to discrimination in various contexts

No Comments

No Known Activations